Comparative Effects of New Generation Oxazaphosphorines on the Size and Viability of Human Acute Myeloblastic Leukemia Cells*
Lidia MAZUR, Ma³gorzata OPYDO-CHANEK, Katarzyna WOJCIESZEK, Marta STOJAK, and Ulf NIEMEYER
Accepted October 5, 2011
MAZUR L., OPYDO-CHANEK M., WOJCIESZEK K., STOJAK M., NIEMEYER U. 2012.
Comparative effects of new generation oxazaphosphorines on the size and viability of human acute myeloblastic leukemia cells. Folia biologica (Kraków) 60: 35-40.
Mafosfamide cyclohexylamine salt (D-17272), 4-hydro-peroxy-cyclophosphamide (D-18864), and$-D-glucose-isophosphoramide mustard (D-19575, glufosfamide) are three new generation oxazaphosphorine agents. The aim of the present study was to compare the cell response to the action of these three oxazaphosphorines. The experiments were performed in vitro on human acute myeloblastic leukemia ML-1 cells. After exposure of ML-1 cells to the oxazaphosphorines, the size, viability and count of these cells were determined. The research was conducted using the spectrophotometric MTT assay and the electronic Beckman Coulter method. The temporary changes in the ML-1 cell size, viability and count, were dependent on the oxazaphosphorine agent tested, its dose, and the time intervals after its application. Among the three oxazaphosphorine agents, D-18864 proved to be the most cytotoxic, and D-19575 was characterized by the lowest cytotoxicity. The results suggest the possibility of using the electronic sizing and counting method and the MTT assay as a rapid in vitro test for assessing leukemic cell sensitivity to the action of new potential chemotherapeutic agents.
Key words: Human leukemic ML-1 cells, oxazaphosphorines, cytotoxicity, cell size, viability, and count.
Lidia MAZUR, Ma³gorzata OPYDO-CHANEK, Katarzyna WOJCIESZEK, Marta STOJAK, Depart- ment of Experimental Hematology, Jagiellonian University, Gronostajowa 9, 30-387 Kraków, Poland.
E-mail: lidia.mazur@uj.edu.pl
malgorzata.opydo-chanek@uj.edu.pl katarzyna.wojcieszek@uj.edu.pl marta.stojak@uj.edu.pl
Ulf NIEMEYER, NIOMECH part of IIT GmbH, University of Bielefeld, Universitätsstr. 25, 33615 Bielefeld, Germany.
E-mail: ulf_niemeyer@t-online.de
Oxazaphosphorines are a class of alkylating agents, widely used in chemotherapy. However, to improve the therapeutic index of this class of chemothera- peutic agents, new generation oxazaphosphorines have been developed. Mafosfamide cyclohexylamine salt (D-17272), 4-hydro-peroxy-cyclophosphamide (D-18864), and $-D-glucose-isophosphoramide mus- tard (D-19575, glufosfamide) are new oxazaphos- phorine derivatives. The cell response to the action of these alkylating compounds is still under ex- perimental and clinical investigation (BODYand YULE2000; ENGELet al. 2000; SEKERet al. 2000;
BLANEYet al. 2005; ZHANGet al. 2005a, 2005b;
LIANGet al. 2007; GIRAUDet al. 2010; MAZURet al.
2011).
The influence of the new generation oxazaphos- phorines on pathological hematopoietic cells is not completely known yet. Available information on the cytotoxic activity of D-17272, D-18864, and D-19575 in human leukemic cells is still scarce (STYCZYÑSKIet al. 2000a, 2000b; GOLDSTEINet al.
2008; MAZURet al. 2009, 2010; OPYDO-CHANEK
et al. 2010a, 2010b).
Therefore, the objective of the present study was to evaluate in vitro cytotoxicity of three new oxazaphosphorine agents on human acute myelo-
_______________________________________
*Supported by Research Projects WRBW/BiNoZ/IZ and K/ZDS/001720.
blastic leukemia ML-1 cells. Temporary changes in the size, viability and count of ML-1 cells sub- jected to the exposure of D-17272, D-18864, and D-19575, were analyzed using the electronic Beckman Coulter method and the MTT assay.
Material and Methods
Cells
Human acute myeloblastic leukemia ML-1 cells (European Collection of Cell Culture, ECACC, Sigma Aldrich, Catalogue No. 88113007) were maintained in RPMI 1640 (Gibco BRL Life Tech- nologies) supplemented with 10% fetal calf serum (GIBCO BRL Life Technologies), 2 mM L-glutamine (Sigma Aldrich), and antibiotic antimycotic solu- tion (AAS, Sigma Aldrich). AAS contained 20 units of penicillin, 20 Fg streptomycin and 0.05 Fg am- photericin B. Leukemic ML-1 cells were passaged every third day. The cells grew expotentially at 37oC in an atmosphere of 5% CO2in air (HERAcell in- cubator, KendroLab). The cultures were periodi- cally tested for Mycoplasma infection.
Chemicals
Mafosfamide cyclohexylamine salt (D-17272, CAS No. 84210-80-0), 4-hydro-peroxy-cyclophosphamide (D-18864, CAS No. 39800-16-3) and glufosfa- mide (D-19575, $-D-glucose-isophosphoramide mustard, CAS No. 132682-98-5) were obtained from NIOMECH (Bielefeld, Germany). D-17272, D-18864 and D-19575 were dissolved in aqua pro injectione (Polpharma). All solutions were freshly prepared directly before treatment of ML-1 cells.
Agent doses and cell treatment
After a dilution of the cell suspension to a den- sity of 15x104cells/ml medium, ML-1 cells were exposed for 30 min to the oxazaphosphorine agent.
D-18864 was applied at a dose of 10 Fg/ml me- dium, D-19575 at a dose of 100 Fg/ml medium, and D-17272 was given at doses of 10 Fg/ml and 100 Fg/ml medium. The control material consisted of untreated ML-1 cells. After 30 min treatment with the oxazaphosphorine agent, leukemic ML-1 cells were centrifuged for 10 min at 1000 rpm and the supernatant was discarded. Then the cells were washed in 2 ml of PBS (BioMed) and pelletted by centrifugation for 7 min. The wash and centrifuga- tion were repeated once more and the cells resus- pended in the complete RPMI 1640 medium.
ML-1 cell analysis
Temporary alterations occurring in human leu- kemic ML-1 cells were assessed at 24h, 48h, and 72h after the cell exposure to the oxazaphosphorine agent. At these three time intervals, the size, vi- ability and count of ML-1 cells, were determined.
Beckman Coulter method of cell sizing and counting
The electronic Beckman Coulter method was ap- plied to determine ML-1 cell size and count. The Beckman Coulter method of cell sizing and count- ing is based on the detection and measurement of changes in electrical resistance produced by a cell suspended in a conductive liquid traversing through a small aperture. The passage of each individual cell momentarily modulates the impedance of the electrical path between two electrodes located on each side of the aperture. An electrical pulse, suit- able for sizing and counting, results from the pas- sage of each cell through the aperture. While the amplitude of the produced electrical pulse depends on the cell volume, the number of pulses indicates the cell count.
Samples of ML-1 cell suspension were taken from flasks and immediately diluted in ISOTON II (Beckman Coulter filtered electrolyte solution based on 0.9 % saline). 500 Fl of the cell suspension was added to 4.5 ml of ISOTON II. After the dilution of ML-1 cell suspension, individual leukemic cells were measured using the Z2 Coulter counter (Beckman Coulter, USA). The count and volume distributions of ML-1 cells were obtained using the counter equipped with a 100 Fm diameter orifice. The flow rate was 500 Fl / 12.5 sec. The range for cell measu- rement was determined as 162-10 370 fL. The cell volume and count was analyzed at 851-10 370 fL.
The instrument was calibrated using 10 Fm diame- ter latex beads (Beckman Coulter CC size stan- dard). The mean cell volume, modal peak and the cell count, were determined using Z2 AccuComp software (Beckman Coulter, USA).
MTT viability assay
The in vitro MTT test was designed for spectro- photometrically determining the viability of cell populations. MTT (3-[4,5-dimethylthiazol-2-yl]
-2,5-diphenyl tetrazolium bromide) is the key component used in the MTT assay. In viable, me- tabolically active cells, the tetrazolium ring is cleaved, yielding formazan crystals. Changes in the metabolic activity of cell populations result in a concomitant change in the amount of formazan formed.
MTT (Sigma-Aldrich) was dissolved in RPMI 1640 medium, at a concentration of 5 mg/ml, and filtered through a 0.2 Fm filter. 100 Fl of the yel- low MTT solution was added to each well of a 24- well plate containing 1 ml of the cell suspension.
The cells were then incubated at 37°C with 5%
CO2. A blank solution was prepared according to the above procedure using complete medium with- out cells. After the three-hour incubation period, the resulting formazan crystals were dissolved with 1 ml of acidified isopropanol (0.05 N HCl in absolute isopropanol), and absorbance of the ob- tained solution was measured at a wavelength of 570 nm using a Pharmacia Ultrospec III spectro- photometer (Pharmacia). The extent of MTT con- version in cells was also expressed as a percentage value of the control.
Statistical evaluation
Statistical significance of differences in the mean cell volume and the mean amount of for- mazan formed as well as in the mean cell count, were evaluated by an analysis of variance and the Duncan’s new multiple range test.
Results
Effects of oxazaphosphorines on ML-1 cells The influence of three new generation oxazaphosphorines on human acute myeloblastic
leukemia ML-1 cells, was determined. The effects of D-18864, D-17272, and D-19575, on the leuke- mic cell size (Table 1 and Table 2, Fig. 1), and their viability (Table 3, Fig. 2) and count (Table 4), ob- served at 24h, 48h, and 72h after the oxazaphos- phorine agent application, were compared.
Mean volume of ML-1 cells (Table 1, Fig. 1) In comparison with the controls, the mean vo- lume of ML-1 cells was increased in all the experi- mental groups and at all three time intervals 24h, 48h, and 72h after the leukemic cell exposure to the action of the oxazaphosphorine agents. Among ML-1 cells treated with the oxazaphosphorine agents, the greatest mean volume was obtained when D-17272 was given at a higher dose of 100 Fg/ml medium, and the smallest mean volume of the leukemic cells was observed after D-19575 ap- plication, at the same dose. The increased mean volume of ML-1 cells exposed to D-17272 at two doses, 10 Fg/ml and 100 Fg/ml medium, appeared to be dose-dependent. The mean volume was found to be smaller in ML-1 cells treated with D-17272, at a dose of 10 Fg/ml medium, than in those exposed to D-18864 at the same dose.
Modal peak of the volume distribution curve (Table 2, Fig. 1)
In relation to the controls, greater values in the modal peaks of the volume distribution curves were found in ML-1 cells exposed to the action of the oxazaphosphorines, except for the value obtained
Table 1 The mean volume of ML-1 cells after their exposure to the oxazaphosphorine agents
Group characteristic Mean cell volume (fL)
24h 48h 72h
No. agent given agent dose mean ± SD mean ± SD mean ± SD
I D-18864 10 Fg/ml 2,3,4,5,48h,72h
3729 ± 75 2,3,4,5,24h,72h
5114 ± 74 2,3,4,5,24h,48h 3926 ± 89
II D-17272 10 Fg/ml 1,3,4,5,48h,72h
3385 ± 51 1,3,4,5,24h,72h
2961 ± 85 1,3,4,5,24h,48h 2440 ± 49
III D-17272 100 Fg/ml 1,2,4,5,48h,72h
3902 ± 62 1,2,4,5,24h,72h
6033 ± 87 1,2,4,5,24h,48h 5238 ± 67
IV D-19575 100 Fg/ml 1,2,3,5,48h,72h
2629 ± 53 1,2,3,5,24h,72h
2819 ± 72 1,2,3,5,24h,48h 2322 ± 90
V control 1,2,3,4,72h
2290 ± 37 1,2,3,4,72h
2309 ± 96 1,2,3,4,24h,48h 2199 ± 91
The data are presented as mean values ± standard deviation.
Statistically significant differences at P<0.05
Differences between groups: different from Group I –1, Group II –2, Group III –3, Group IV –4, Group V –5. Differences within each group: different from 24h –24h, 48h –48h, 72h –72h.
at 72h after D-17272 application at a higher dose of 100 Fg/ml medium. The patterns of temporary changes in the modal peak of the volume distribu- tion curve are correlated with those of the mean volume of ML-1 cells.
ML-1 cell viability (Table 3, Fig. 2)
The cell viability is correlated with the optical density of formazan solution. Compared with the controls, the optical density of formazan solution decreased 24h, 48h, and 72h after the treatment of ML-1 cells with D-18864, D-17272, and at 24h and 72h after their exposure to the action of D-19575. Among ML-1 cells treated with the oxazaphosphorine agents, the values of the optical density of formazan solution were found to be greater following D-19575 application at a dose of 100 Fg/ml, and when D-17272 was given at a dose of 10 Fg/ml medium, and smaller values were ob- tained when D-18864 was applied at a dose of 10 Fg/ml and D-17272 at a dose of 100 g/ml medium.
The viability rate indicating the extent of MTT conversion to formazan, and expressed as the per- centage values of the controls, appeared to be dis- tinctly reduced in ML-1 cells exposed to D-18864 at a dose of 10 Fg/ml and especially in those treated with D-17272 at a dose of 100 Fg/ml medium.
ML-1 cell count (Table 4)
In comparison with the controls, the cell count of ML-1 cells decreased in all experimental groups and in all three time intervals at 24h, 48h, and 72h after the leukemic cell exposure to the action of the oxazaphosphorine agents. Among ML-1 cells treated with the oxazaphosphorine agents, the smallest cell number was found when D-17272 was given at a higher dose of 100 Fg/ml medium, and the greatest number of the leukemic cells was encountered after D-19575 application, at the same dose. A reduced
Fig. 1. The volume of ML-1 cells exposed to the oxazaphosphorines.
The volume distribution curves of ML-1 cells analyzed with the Beckman Coulter counter. The peaks on the left represent cellular debris, presumably apoptotic bodies and necrotic cell fragments, which were excluded from the analysis of the ML-1 cell volume.
Table 2 The modal peaks of the volume distribution curves of ML-1 cells exposed to the action of D-17272, D-18864, and D-19575
Group characteristic Modal peak (fL)
No. agent given agent dose 24h 48h 72h
I D-18864 10 Fg/ml 3181 3628 3222
II D-17272 10 Fg/ml 2735 2290 1925
III D-17272 100 Fg/ml 3506 7437 871
IV D-19575 100 Fg/ml 1965 2168 1804
V control 1682 1804 1601
count of ML-1 cells exposed to D-17272 at two doses, 10 Fg/ml and 100 Fg/ml medium, appeared to be dose-dependent. The cell number was ob- served to be greater in ML-1 cells treated with D-17272, at a dose of 10 Fg/ml medium, than in those exposed to D-18864 at the same dose.
Discussion
The results of the present study have shown the human acute myeloblastic leukemia ML-1 cell re- sponse to the action of three new generation oxazaphosphorines D-17272, D-18864, and D-19575.
Temporary alterations in the size, viability and count of ML-1 cells, observed 24h, 48h and 72h
Fig. 2. Effects of the oxazaphosphorines on the viability rate of ML-1 cells. The extent of MTT conversion to formazan in ML-1 cells expressed as a percentage value of the control. D-18864/10 - D-18864 applied at a dose of 10 Fg/ml medium; D-17272/10 - D-17272 applied at a dose of 10 Fg/ml medium; D-17272/100 - D-17272 applied at a dose of 100 Fg/ml medium; D-19575/100 - D-19575 applied at a dose of 100 Fg/ml medium.
Table 3 The optical density of formazan solution determined after ML-1 cell exposure to the oxazaphosphorine agents
Group characteristic Optical density of formazan solution (x10-1)
24h 48h 72h
No. agent given agent dose mean ± SD mean ± SD mean ± SD
I D-18864 10 Fg/ml 2,4,5,48h,72h
2.13 ± 0.08 2,3,4,5,24h
3.16 ± 0.07 2,3,4,5,24h 3.09 ±0.09
II D-17272 10 Fg/ml 1,3,5,48h,72h
2.26 ± 0.13 1,3,5,24h,72h
4.24 ± 0.14 1,3,4,5,24h,48h 4.92 ± 0.25
III D-17272 100 Fg/ml 2,4,5,48h,72h
2.13 ± 0.05 1,2,4,5,24h
2.92 ± 0.06 1,2,4,5,24h 2.86 ± 0.10
IV D-19575 100 Fg/ml 1,3,5,48h,72h
2.45 ± 0.14 1,3,24h,72h
4.27 ± 0.15 1,2,3,5,24h,48h 5.13 ± 0.12
V control 1,2,3,4,48h,72h
2.57 ± 0.07 1,2,3,24h,72h
4.33 ± 0.17 1,2,3,4,24h,48h 5.49 ± 0.09
For explanation of signs, see Table 1.
Table 4 The count of ML-1 cells after their exposure to the oxazaphosphorine agents
Group characteristic Cell count (x 103)
24h 48h 72h
No. agent given agent dose mean ± SD mean ± SD mean ± SD
I D-18864 10 Fg/ml 2,4,5,48h,72h
169.99 ± 5.06 2,3,4,5,24h,72h
221.18 ± 5.75 2,3,4,5,24h,48h 300.76 ± 8.18
II D-17272 10 Fg/ml 1,3,4,5,48h,72h
225.88 ± 6.40 1,3,4,5,24h,72h
632.50 ± 12.97 1,3,4,5,24h,48h 1049.91 ± 14.08 III D-17272 100 Fg/ml 159.03 ± 3.832,4,5,72h 1,2,4,5,72h
155.59 ± 5.86 1,2,4,5,24h,48h 134.62 ± 4.57
IV D-19575 100 Fg/ml 1,2,3,5,48h,72h
281.39 ± 10.50 1,2,3,5,24h,72h
704.49 ± 38.92 1,2,3,5,24h,48h 1150.73 ± 79.70
V control 1,2,3,4,48h,72h
374.63 ± 14.07 1,2,3,4,24h,72h
818.78 ± 63.04 1,2,3,4,24h,48h 1373.67 ± 15.68
For explanation of signs, see Table 1.
after their exposure to the oxazaphosphorines agents, were dependent on the agent tested, its dose, and the time intervals after its application.
Among the oxazaphosphorine agents, D-18864 appeared to be the most cytotoxic, and D-19575 was characterized by the lowest cytotoxicity. The variable leukemic cell responses to the action of the oxazaphosphorines, were found.
The cytotoxic effects of the oxazaphosphorines, D-17272, D-18864, and D-19575, surely resulted from their various action on human leukemic ML-1 cells. However, the precise mechanisms re- sponsible for cytotoxicity of these three oxazaphosphorine agents on pathological hemato- poietic cells have not yet been completely eluci- dated. It is accepted that phosphoramide mustard and acrolein are the major reactive alkylating agents of D-17272 and D-18864. D-19575 con- tains the direct alkylating moiety, isophosphora- mide mustard. The reactive alkylating compounds are involved in the toxicologic reactions because they can bind to a variety of cellular molecules dis- turbing the normal processes occurring in the cell.
Moreover, the cytotoxicity of these oxazaphos- phorines is dependent on their pharmacokinetic and pharmacodynamic properties (BODYand YULE
2000; ZHANGet al., 2005a, 2005b; LIANG et al.
2007; GIRAUDet al. 2010; MAZURet al. 2011).
The changes in the metabolic activity and cell vi- ability, the volume and count of leukemic ML-1 cells following D-17272, D-18864, and D-19575 application can result from various alterations oc- curring in these cells at the molecular, biochemical and morphological levels. In the previous publica- tion it has been shown that damage caused by the oxazaphosphorine agents in the hematopoietic cells was manifested as DNA breakage, cell cycle disruption, cell proliferation inhibition, mitotic ca- tastrophe and programmed cell death – induction (MAZUR2002; MAZURet al. 2009, 2010; OPYDO- -CHANEKet al. 2010a, 2010b).
To summarize, the results of the present investi- gations indicated that three analyzed parameters, cell volume, viability, and cell count are important in determining and comparing cytotoxic effects observed in human leukemic cells after their expo- sure to the new generation oxazaphosphorines. In conclusion, the electronic sizing and counting method and MTT assay can be used to study the anticancer activity of newly synthesized and tested, potential chemotherapeutic agents.
Acknowledgements
The authors wish to thank Urszula K£APUTfor her excellent technical assistance.
References
BLANEY S. M., BALISF. M., BERG S., ARNDT C. A. S., HEIDEMANR., GEYERJ. R., PACKERR., ADAMSONP. C., JAECKLE K., KLENKE R., AIKIN A., MURPHY R., MC CULLYC., POPLACKD. G. 2005. Inthrathecal mafosfamide:
Preclinical pharmacology and phase I trial. J. Clin. Oncol.
23: 1555-1563.
BODYA. V., YULES. M. 2000. Metabolism and pharmacoki- netics of oxazaphosphorines. Clin. Pharmakokinet. 38:
291-304.
ENGELJ., KLENNERT., NIEMEYERU., PETERG., POHLJ., SCHÜßLERM., SCHUPKEA., VOSSA., WEISSLERM. 2000.
Glufosfamide. Drug. Future 25: 791-794.
GIRAUD B., HEBERT G., DEROUSSENT A., VEAL G. L., VASSAL G., PACIA. 2010. Oxazaphosphorines: new thera- peutic strategies for an old class of drugs. Expert. Opin. Drug Metab. Toxicol. 6: 919-938.
GOLDSTEINM., ROOSW. P., KAINAB. 2008. Apoptotic death induced by the cyclophosphamide analogue mafosfamide in human lymphoblastoid cells: Contribution of DNA replica- tion, transcription inhibition and Chk/p53 signaling. Toxi- col. Appl. Pharmacol. 229: 20-32.
LIANGJ., HUANGM., DUANW., YUX. O., ZHOUS. 2007. De- sign of new oxazaphosphorine anticancer drugs. Curr.
Pharm. Des. 13: 963-978.
MAZURL. 2002. Indukowanie apoptozy i nekrozy komórek hematopoetycznych przez tiole, promieniowanie jonizuj¹ce i zwi¹zki chemioterapeutyczne (Induction of apoptosis and necrosis of hematopoietic cells by thiols, ionizing radiation and chemotherapeutic agents). Kraków, Ksiêgarnia Akade- micka, 1-120. (In Polish).
MAZURL., OPYDO-CHANEKM., STOJAKM. 2011. Glufosfa- mide as a new oxazaphosphorine anticancer agent. Anti- Cancer Drug. 22: 488-493.
MAZUR L., OPYDO-CHANEK M., STOJAK M., BARAN J., NIEMEYERU. 2010. Induction of DNA breakage in U937 cells by oxazaphosphorines. Folia biol. (Kraków) 58: 15-20.
MAZURL., STOJAKM., OPYDO-CHANEKM., NIEMEYERU.
2009. Mitotic catastrophe induction in U937 cells by oxazaphosphorines. Acta Biol. Cracov. Ser. Zool. 51: 17-22.
OPYDO-CHANEKM., MAZURL., STOJAKM., NIEMEYERU.
2010a. Comparative effects of new generation oxazaphos- phorines on cancer cell death. 6th Swiss Apoptosis Meeting, Bern, Switherland, 46.
OPYDO-CHANEKM., STOJAKM., MAZURL., NIEMEYERU.
2010b. Changes in the size of U937 cells following exposure to new generation oxazaphosphorines. Acta Biol. Cracov.
Ser. Zool. 52: 25-29.
SEKERH., BERTRAMB., BURKLEA., KAINAB., POHLJ., KOEPSELLH., WIESSLERM. 2000. Mechanistic aspects of the cytotoxic activity of glufosfamide, a new tumor thera- peutic agent. Br. J. Cancer 82: 629-634.
STYCZYÑSKIJ., WYSOCKIM., DÊBSKIR., BALWIERZW., ROKICKA-MILEWSKAR., MATYSIAKM., BALCERSKAA., KOWALCZYKJ., WACHOWIAKJ., SOÑTA-JAKIMCZYKD., CHYBICKAA. 2002a. In vitro activity of oxazaphosphorines in childhood acute leukemia: Preliminary report. Acta Bio- chim. Pol. 49: 221-225.
STYCZYÑSKIJ., WYSOCKIM., KURYLAKA., JURASZEWSKA E., MALINOWSKA I., STANCZAK E., P£OSZYÑSKA A., STEFANIAK J., MAZUR B., SZCZEPAÑJSKI T., RAS M.
2000b. In vitro activity of glufosfamide in childhood acute leukemia. Anticancer Res. 22: 247-250.
ZHANGJ., TIANQ., CHANS.Y., DUANW., ZHOUS. 2005a.
Insight into oxazaphosphorine resistance and possible ap- proaches to its circumvention. Drug Resist. Uptake 8:
271-297.
ZHANGJ., TIANQ., CHANS. Y., LIS. C., ZHOUS. 2005b. Me- tabolism and transport of oxazaphosphorines and the clinical implications. Drug Metab. Rev. 37: 611-703.
